Você está na página 1de 52

OEG 001 ROOT CAUSE EVALUATION

Wisconsin Electric
Power Company
DOCUMENT TYPE: Controlled Reference

REVISION: 3

EFFECTIVE DATE: April 29, 1999

PROCEDURE OWNER: Performance Assessment and Corrective Action

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION TABLE OF CONTENTS SECTION 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 EXHIBIT A EXHIBIT B EXHIBIT C EXHIBIT D EXHIBIT E EXHIBIT F EXHIBIT G EXHIBIT H TITLE

OEG 001 Revision 3 April 29, 1999

PAGE

PURPOSE ...................................................................................................................... 4 REFERENCES............................................................................................................... 4 DEFINITIONS ............................................................................................................... 4 PRECAUTIONS ............................................................................................................ 6 PREREQUISITES.......................................................................................................... 6 PROCESS ...................................................................................................................... 6 REPORT PREPARATION ........................................................................................... 13 ENTRY OF ACTION ITEMS INTO NUTRK ............................................................. 14 EQUIPMENT FAILURE ROOT CAUSE.................................................................... 15 COMMON CAUSE ANALYSIS (CCA)...................................................................... 16 CORRECTIVE ACTION REVIEW BOARD (CARB)................................................ 17 EFFECTIVENESS REVIEWS ..................................................................................... 17 PERSONAL STATEMENT ......................................................................................... 18 EVENT AND CAUSAL FACTOR CHARTING......................................................... 19 TASK ANALYSIS........................................................................................................ 23 INTERVIEWING.......................................................................................................... 25 CHANGE ANALYSIS ................................................................................................. 30 BARRIER ANALYSIS................................................................................................. 32 FAILURE MODES AND EFFECTS ANALYSIS....................................................... 35 CAUSE AND EFFECT ANALYSIS............................................................................ 38

Page 2 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION TABLE OF CONTENTS SECTION EXHIBIT I EXHIBIT J EXHIBIT K EXHIBIT L TITLE

OEG 001 Revision 3 April 29, 1999

PAGE

PARETO ANALYSIS................................................................................................... 39 TROUBLESHOOTING / FAILURE ANALYSIS ....................................................... 40 FAULT TREE ANALYSIS .......................................................................................... 41 DEVELOPMENT OF RECOMMENDATIONS FOR CORRECTIVE ACTIONS .... 42

EXHIBIT M EVENT EVALUATION AND ROOT CAUSE ANALYSIS TECHNIQUES APPLICATION GUIDELINE ............................................................ 43 EXHIBIT N EXHIBIT O EXHIBIT P EXHIBIT Q EXHIBIT R RCE QUALITY INDEX SCORE SHEET EXAMPLE................................................ 44 ROOT CAUSE EVALUATION PLANNING GUIDE ................................................ 46 ROOT CAUSE ANALYSIS PACKAGE COMPLETION CHECKLIST ................... 47 EFFECTIVENESS REVIEWS ..................................................................................... 49 APPARENT CAUSE DETERMINATION GUIDELINES ......................................... 51

Page 3 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

1.0

PURPOSE The purpose of this document is to provide guidance for personnel to effectively identify the root cause(s) of problems to ensure proper corrective actions to prevent recurrence are implemented. This document provides guidance for an investigator to determine a root cause of an event. It is the investigators responsibility to select the most appropriate analysis technique, whether covered by this guide or not, that will identify the root cause(s).

2.0

REFERENCES 2.1 2.2 2.3 2.4 2.5 INPO 90-004, "Good Practice OE-207, Root Cause Analysis" NP 5.3.1, "Condition Reporting System" OM 4.1.1, "Post-Trip Review" ECL-5, "Post-Trip Reviews" NP 5.4.1, "Open Item Tracking Systems"

3.0

DEFINITIONS 3.1 Apparent Cause Evaluation (ACE): An evaluation done to determine the apparent cause (versus a root cause) of an event. An apparent cause evaluation is less rigorous than a root cause evaluation. (See Exhibit R) Causal Factors: The potentially influencing conditions or elements that were present when a condition adverse to quality occurred that may have led to or contributed to the root or contributing cause(s). Corrective Action (CA): Action taken to restore the adverse condition to an acceptable condition or capability (Full Qualification), but may not be the only actions needed to prevent recurrence of the condition. Interim Corrective Actions (ICA): Actions taken to reduce the risk of reoccurrence while waiting for CAPR. Corrective Action to Prevent Recurrence (CAPR): Action taken to prevent recurrence of a condition or event or to eliminate or minimize the causal factors of the condition. Contributing Cause: Causes that, if corrected would not by themselves have prevented the event, but are important enough to be recognized as needing corrective action to improve the quality of the process or product.

3.2

3.3

3.4

3.5

3.6

Page 4 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 3.7

OEG 001 Revision 3 April 29, 1999

Immediate Action: Action taken to promptly restore a condition adverse to quality to an acceptable state. After evaluation, additional prompt actions may be required to fully restore the deficient condition. Root Cause: Identified cause(s) that, if corrected, will prevent recurrence of a condition adverse to quality. Root Cause Evaluation (RCE): Evaluation that determines and analyzes the cause(s) of an inappropriate action or adverse condition, then identifies the Root Cause(s) of the event. Root Cause Investigator (RCI): A qualified individual assigned by the Line Organization to perform a root cause evaluation. SCAQ Action: A corrective action that serves to eliminate the root(s) or significant contributing cause(s) of a Significant Condition Adverse to Quality. Failure Mode: An event causal factor that when identified will help identify the Root Cause(s) and Contributing Cause(s) for an event. Common Cause Assessment (CCA): An assessment method used to identify the Root Cause(s) and Contributing Cause(s) for a number of similar events. Usually initiated based on a declining or adverse trend, the analysis generally uses a variety of statistical analyses, interviews, and surveys to help to determine the Root Cause(s) of the adverse trend. Combined Root Cause Evaluation (RCE): More than one apparently similar event is analyzed in one RCE report. Evaluation determines and analyzes the apparent cause(s) of an inappropriate action or adverse condition for each report, then identifies the Root Cause(s) of the events. Analysis not as extensive as CCA. Equipment Failure Root Cause Evaluation (RCE): An assessment of equipment failures where the failure modes are the result of material, design, or similar equipment-related defects or natural phenomenon (e.g., tornado, lightning). This should include Maintenance Rule failures and should consider Human Error or Organizational/Programmatic Breakdown failure modes.

3.8

3.9

3.10

3.11

3.12

3.13

3.14

3.15

Page 5 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 4.0 PRECAUTIONS 4.1

OEG 001 Revision 3 April 29, 1999

Preservation of physical evidence and important information is necessary to determine root causes. Investigators should plan activities so that physical evidence and other important information is not altered, destroyed, or lost. Preservation of evidence must not interfere with or delay placing the plant or systems in a safe condition. The root cause investigator must not become distracted by event recovery activities. Investigators should communicate effectively with recovery team members, but stay focused on investigation and root cause analysis. A root cause investigator should refer to this guide as appropriate, while performing evaluations. The intent of the guide is to improve the efficiency and effectiveness of evaluations. A Root Cause Evaluation (RCE) requires a parent NUTRK document to track corrective actions.

4.2

4.3

4.4

5.0

PREREQUISITES If a reactor trip has occurred, operations personnel will have obtained initial information and statements per PBNP OM 4.1.1 and the associated ECL-5 checklist. Obtain copies of this information to support your evaluation.

6.0

PROCESS 6.1 Preparation Initiate the preparation process as soon as practicable after the evaluation is assigned. The following points should be helpful to the investigator to better plan the evaluation. Determine the scope of the evaluation with the appropriate line manager. When planning the evaluation, consider who should be interviewed and any schedule constraints that may impact the interviews (e.g., shift workers). If support from another department is involved, give them early notification. Give early consideration to the need to correspond with outside organizations such as vendors, EPRI, other utilities, etc., if needed to support the evaluation. Sometimes information requests and inquiry responses can take several days or weeks. NOMIS and Nuclear Network are two industry information exchange media for requesting information from other utilities who may have experienced similar events. Identify or define the station acceptable performance criteria that meets or exceeds applicable Industry Standards and Regulations. If performing an RCE on an incident that involves chemicals or chemical processes, contact Industrial Health and Safety to ensure compliance with OSHA 1910.

Page 6 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 6.2

OEG 001 Revision 3 April 29, 1999

Exhibit O provides an example of a planning checklist that can be used to work out the schedule of completing the RCE. Approximate average industry time for completion of a RCE is as follows: Estimated Man-hours: Common Cause = 100 - 700 (Hours may vary greatly based on extent of problem/size of team). Root Cause = 40 to 80 (significant management review and revision may extend this). Apparent Cause = 1-10 (used when causes and corrective actions are readily apparent).

6.3

Information Gathering 6.3.1 The investigator should gather information and data relating to the event/problem. This includes physical evidence, interviews, records, and documents needed to support the root cause. Some typical sources of information which may be of assistance include the following: Operating logs Maintenance records Inspection reports Procedures and Instructions Vendor Manuals Drawings and Specifications Equipment History Records Strip Chart Recordings Trend Chart Recordings Sequence of Event Recorders Radiological Surveys Plant Parameter Readings Sample Analysis and Results Correspondence Design Basis Information Photographs/Sketches of Failure Site Industry Bulletins Previous CRs/Root Cause Evaluation Reports (RCEs) NPRDS Records Turnover logs for affected groups (e.g., HP, Maintenance) Task sheets Lesson plans NUTRK Trend Data, including OE Narrow Issue Trending Data.

Page 7 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION NOTE:

OEG 001 Revision 3 April 29, 1999

Statements should be obtained prior to any critique which could alter the perceptions of those involved whenever possible. Use Exhibit A, Personnel Statement, or a similar form to obtain written statements from personnel involved as soon as practical (preferably prior to leaving the site) following the event. Personnel statements are normally written separately by each individual rather than as a collaborative summary of the event. Construction of an Event and Causal Factor Chart should begin as soon as information becomes available. Even though the initial event sequence and timeline may be incomplete, it should be started early in the evaluation process. Construct an Event and Causal Factor Chart that shows the order in which each action of the event occurred. This can most easily be done by compiling all input information (e.g., interviews, written statements, evaluation results) and placing them in chronological order. A Task Analysis may be useful in constructing the Event and Causal Factor Chart. See Exhibit B, Event and Causal Factor Charting, and Exhibit C, Task Analysis. Conduct personnel interviews with involved parties as soon as practical following the event. See Exhibit D, Interviewing. If it is suspected that the cause of the event may have been an intentional attempt to disrupt normal plant operation (e.g., tampering), notify Station and Nuclear Operations management and the Director of Corporate Security in accordance with applicable station procedures.

6.3.2

NOTE:

6.3.3

6.3.4

6.3.5

6.4

Analyzing Information NOTE: These are not the only methods available, but represent proven techniques for evaluating various types of problems. Using the facts identified by the evaluation, and reviewing the event as a whole, decide which of the facts or groups of facts are pertinent. Analytical techniques that may be helpful include: Change Analysis (Exhibit E) Barrier Analysis (Exhibit F) Failure Modes and Effects Analysis (Exhibit G) Cause and Effects Analysis (Exhibit H) Pareto Analysis (Exhibit I) Troubleshooting/Failure Analysis (Exhibit J) Fault Tree Analysis (Exhibit K)

6.4.1

Page 8 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 6.4.2

OEG 001 Revision 3 April 29, 1999

Compare the facts to an "acceptable standard" and determine if an unacceptable condition exists. Identify each inappropriate action and equipment failure. Review the PBNP RCE list and identify previous and current investigations that may influence your evaluation or your corrective actions. Search NUTRK for key words or similar What, Who, and Why codes that could identify other related issues, past or present. Review the corrective actions from these other events and determine how effective they were in preventing or mitigating recurrence of the event. If youre reading this note you should take 5 minutes to contact the Corrective Action Program Group and have them establish your Nuclear Network access. They can show you how to conduct a search. No evaluator should be without it! The Nuclear Network can be used to identify similar events or OE information. Searches performed using this network are easy and efficient. Why not identify industry standards or validate your proposed solutions against proven programs at other sites? If you still cant find answers just place your question on the network and watch how fast the system works. The OE Coordinator can also help with searches. Review the corrective actions from other events or OE evaluations and determine how effective they were in preventing the recurrence or mitigating the outcome of the current event. Consider whether any corrective actions still in progress could have prevented the event or mitigated the outcome of the event. All RCEs should address "EXTENT OF CONDITION." Ask the question, "Could this condition be lurking out there some where else?" If it is truly isolated and not applicable to anything else, state it explicitly in your report. Otherwise we need to determine the extent of the condition or how we will determine the extent. An easy example would be a CR on electronics data in tables being corrupted. Clearly, it could affect all procedures which had been converted from Word Perfect to Microsoft Word. We are going to look at them all to see if there are others with problems. A harder example would be the recent RCE on issues with the Aux Feed not being able to take a single failure after a tornado. Until recently, we didnt consider this to be part of our design basis. Are there other systems out there that could be susceptible to single failures after tornadoes/seismic events? Ensure similar components or documents are examined to determine the extent to which the unacceptable condition exists.

6.4.3

NOTE:

6.4.4

6.4.5

NOTE:

6.4.6

Page 9 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 6.4.7 6.4.8 6.4.9 6.4.10

OEG 001 Revision 3 April 29, 1999

Evaluate potential detrimental effects on associated plant equipment. Organize the information into an overall description of the problem. Establish a start time and a finite end time to the event. Occasionally, more than one apparently similar event is analyzed in one RCE report (often to reduce backlog). The evaluation should use the analysis techniques described above to determine and analyze the pertinent facts, extent of condition, failure mode(s), etc., of the inappropriate action or adverse condition for each event or issue, then identify the root cause(s). Each event needs to be considered separately first as the causes may actually not be related at all (for example, three storage tanks failing over the course of a month may sound similar with a potential common root cause, but one might be due to a system lineup causing overpressurization, one due to a tornado, one due to corrosion). It is important to ensure that all issues and corrective actions required by the individual CRs or RCEs are addressed in the final report.

6.5

Root Cause Determination 6.5.1 Once the Event and Causal Factor Chart has been constructed, it may be necessary to break down the sequence of events further to determine causal and contributing factors that led to each inappropriate action or equipment failure. Root cause(s) will be determined from the causal factors. The PII Charts with codes are located in Electronic file; J:\DATA\NP\OE\OEGUIDES\ROOTCAUS\PIIChart.doc The failure modes (causal factors) should be determined by using the PII Executive Management Failure Mode Chart (EMFMC), Organizational & Programmatic Diagnostic Chart (OPDC), and Human Error Failure Mode Charts (HEFMC). Each failure mode must be supported by facts determined in the investigation. Not all facts may necessarily lead to a failure mode; also, multiple facts may lead to a single failure mode and individual facts may lead to multiple failure modes. Organizational & Programmatic (O&P) issues may initially be identified during interviews, but the issues should be verifiable through factual information such as procedures, process maps (PII OPIC charts), regulations, etc.

NOTE:

6.5.2

6.5.3

Page 10 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION NOTE:

OEG 001 Revision 3 April 29, 1999

Normally, more than one failure mode is involved with an event. The failure mode is not a Root Cause, but a means to help determine the root cause(s). Once all the failure modes are identified determine the potential Causes by stream analysis. Using a copy of the PII chart, for each failure mode identified, draw lines to the other failure modes that it caused; then draw lines to the failure mode from each of the others that it was caused by. When all cause-effect relationships have been identified, count how many lines go out from and into each box on the chart. Failure modes with the most lines going out are causes, the ones with the most coming in are effects (although they may also be causes); the failure mode with the most should be related to the root cause. This is a graphical analysis similar to the analysis in the next step. For each causal factor identified, ask the following questions until the root cause(s) is determined (see Exhibit H, Cause and Effect Analysis). What caused this? Why does this condition exist?

6.5.4

6.5.5

6.6

Root Cause Determination and Validation 6.6.1 Once the causes of an event have been identified, take action to ensure that the correction of the causes will prevent recurrence. If a cause does not meet all three of the required criteria but meets 1 or 2, then it is considered a "significant contributing" cause. Each root cause should meet the following three criteria: The problem would not have occurred had this cause not been present. The problem will not recur due to the same cause if it is corrected or eliminated. Correction or elimination of the cause(s) will prevent recurrence of similar conditions.

NOTE:

6.6.2

Page 11 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION NOTE:

OEG 001 Revision 3 April 29, 1999

Solutions will often require the implementation of various actions depending on required plant conditions or resource availability Corrective Action (CA)- action taken to restore full qualification. Interim Corrective Actions (ICA)- actions taken to reduce the risk of reoccurrence while waiting for long term CAPR. Corrective Action to Prevent Reoccurrence (CAPR) - actions taken to preclude reoccurrence of the adverse condition.

6.7

Corrective Action Development 6.7.1 Solutions must be identified and implemented that will correct the identified root cause(s) Brainstorming, and interviewing are good sources of ICAs/CAPRs and involve people to establish ownership as early as possible. See Exhibit L, Development of Recommendations For Corrective Actions. Apply the following criteria to CAPRs to ensure they are viable. 6.7.4 Will these CAPRs prevent recurrence of the problem? Are the CAPRs within the capability of Point Beach to implement in a cost effective manner? Do the CAPRs allow Point Beach to meet its primary objectives-of safety and consistent electrical generation? Will the implementation of the CAPRs result in meeting or exceeding applicable industry standards.

6.7.2

6.7.3

If the investigator, sponsor, or a group responsible for implementing corrective actions is unable to reach agreement, the CAP Manager will facilitate a resolution. When necessary, CARB will provide the final resolution. Investigators assist Groups in establishing the implementation conditions and resources. Accurate due dates (completion dates), require adequate planning and dedication of group resources. Your report must reflect this commitment. When appropriate identify ICA(s) which will be implemented while waiting for CAPR(s).

6.7.5

Page 12 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 6.7.6

OEG 001 Revision 3 April 29, 1999

Corrective actions should be specific and address each cause. Corrective actions too narrowly focused, unless appropriate, may not correct the root cause such that similar events are prevented. Corrective actions that are too broad or more extensive than the causes would imply may be an attempt to shotgun the cause or to use the RCE to correct pet peeves. In the long run, this could prove costly and create the potential for another event. The PII Charts with codes are located in Electronic file; J:\DATA\NP\OE\OEGUIDES\ROOTCAUS\PIIChart.doc The PLA shall update Trend Data in NUTRK (NP 5.3.1) to reflect the actual when, who, why, what and system codes.

NOTE:

6.7.7

7.0

REPORT PREPARATION NOTE: You can find a template for the RCE report format at J:\DATA\NP\OE\RCE\TEMPLATE\Rce.dot.

7.1

The RCE report should contain: a cover page which is marked "Company Confidential," title, the CR/QCR and LER (if appropriate) associated with the event, and the names of the investigator(s). an executive summary which includes a brief summary of the event, its consequences and significant corrective actions. Limit the summary to one page. an events and casual factors chart. an event narrative. extent of the adverse condition. a summary of root and contributing causes and proposed corrective actions. an investigators' comments section, if appropriate. recommended corrective actions with responsible group and completion date. Corrective Actions to Prevent Recurrence should be identified. a header containing "Company Confidential," title and RCE number should be on all pages. discussion of equipment function and licensing basis, if an equipment failure root cause is being performed.

The summary of causes and proposed corrective actions will identify the "root" and "significant contributing" cause(s) and a brief explanation as to how it relates to the condition or event. Each of these causes should have an associated SCAQ related corrective action. 7.2 After the RCE report is drafted, the cognizant managers and potentially affected personnel should be given the opportunity to review and comment on the draft report in its entirety prior to finalization.

Page 13 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 7.3

OEG 001 Revision 3 April 29, 1999

The draft of the RCE should be given to the RCE Coordinator prior to final approval for review in accordance with the Root Cause Analysis Quality Index (RCQAI). This index assesses RCE quality according to a set of questions that can be compared against industry results. This index will produce a score for the RCE and provides for feedback of specific comments to the RCI. Any RCE not scored prior to approval will be scored after approval and distribution. (See Exhibit N) The cover page of the RCE will contain the signature and date of the Issue Manager, Corrective Action Manager, and when directed by the CAP Manager, other managers that may be significantly affected. When the RCE has been approved by the appropriate managers, it is routed to the RCE Coordinator. The RCE Coordinator shall review the RCE and NUTRK to ensure corrective actions have been entered. If corrective actions have not been entered, the RCE Coordinator will notify the PLA that the RCE is approved and corrective actions should be entered. The RCE Coordinator shall distribute the evaluation to all NPBU Section Heads, Off-Site Review Committee, and, depending on the event issue, other appropriate personnel. Entry of corrective actions into NUTRK is independent of CARB review.

7.4

7.5

7.6

NOTE: 7.7

The RCE is considered approved when the appropriate managers sign the cover sheet. As soon as the RCE is approved, close the NUTRK evaluation associated with the RCE and notify the PLA to enter corrective actions in NUTRK.

8.0

ENTRY OF ACTION ITEMS INTO NUTRK CR/QCR PLA will ensure that corrective actions for "root" and "significant contributing" causes are tracked through NUTRK Action Items (Child Records) generated under the SCAQ Parent Record. It is the PLAs responsibility to ensure that CATPR are properly identified in NUTRK.

Page 14 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 9.0 EQUIPMENT FAILURE ROOT CAUSE.

OEG 001 Revision 3 April 29, 1999

The depth to which the equipment failure root cause analysis is taken is based on the safety and economic significance of the failure. If root cause analysis is required, then full analysis should be conducted. See Exhibit M for possible techniques to use. 1. Quarantine or preserve the failed equipment so that evidence is not destroyed or disturbed. 2. Determine potential failure modes. 3. Develop a testing plan that utilizes the failure modes chart to prove or refute all the possible causes. The testing plan should prevent destruction of evidence as much as possible for future testing and should detail the expected resulting possibilities. 4. Through the testing sequence different failure modes should be eliminated. The goal is to eliminate all but one failure mode. The failure mode should determine the root cause. 5. If testing shows that multiple event failure modes have taken place in the same event, then each must be considered for root cause and corrective actions should be applied to each unique root cause. 6. In these evaluations, the following additional items should be considered during the investigation. Findings in each of the items below must be provided in the final report: Current Operability Determination or evaluation. Reportability evaluation. Transportability evaluation. Industry Operating Experience (OE) review. Internal Operating Experience (OE) review. Vendor experience/input. Organizational & Programmatic Deficiencies/Human Error contribution.

7. Investigation of the failure mode may require laboratory analysis. Many of these test results must be compared to the original design specifications to determine if the critical characteristics of the failed item meet design requirements. Tolerances should be included as this will often identify a mis-manufactured item. 8. Successful equipment failure root cause is heavily dependent on a thorough and systematic evaluation of technical data. After collecting the data, perform simple analyses to eliminate possible scenarios. Watch for human error or programmatic problems. Consult experts as required. 9. In the case of less significant equipment failures, an apparent cause analysis should be performed. Preventive measures should not normally be prescribed as the root cause may not be known with adequate certainty. Remedial corrective actions, such as repair or replace failed components, should be provided. Page 15 of 52 INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 10.0 COMMON CAUSE ANALYSIS (CCA) 1. Data Collection (from CRs, LERs, NOVs, etc.)

OEG 001 Revision 3 April 29, 1999

Data is generally from cause coding from completed evaluations. If any evaluation is still open that is to be included as part of the data, at a minimum an apparent cause needs to be determined and coded for analysis. Data may need to be transferred to another application for generating charts used in analysis. Data may need to be validated and possibly recoded due to variations on how people code similar events. 2. Develop process chart (PII OPIC chart) Identify key activities (e.g., request work, prepare work plan, etc.) Chart should have just 15-20 elements When coding key activity, relate to chart The key activities will direct what questions need to be asked

3. Categorize Data, typically in six key areas (minimum): Organization(s) Work Process(es) Key Activity Organizational/Programmatic Failure Mode Human Error/Inappropriate Action Failure Mode Human Error Type (Skill, Rule, or Knowledge-Based)

4. Plot Nomographs/Pareto Charts (see Exhibit I) The error rate of the data plotted will determine which information in the charts is not used in further analysis (i.e., insignificant). The more data used, the smaller the error rate. Generally, for 50 bits of data, look at patterns above 8%; for 100 bits, above 6%. For a general common cause analysis, patterns above two times the error rate are often considered; for a process specific CCA, one times the error rate is appropriate.

5. Analyze for Common Causes or Common Characteristics or Common Failure Modes. Primary Failure Modes would be determined using Stream Analysis. 6. Perform further Root Cause and Quantitative Analysis as appropriate.

Page 16 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION 11.0 CORRECTIVE ACTION REVIEW BOARD (CARB)

OEG 001 Revision 3 April 29, 1999

Corrective Action Review Board (CARB) reviews all Level A and Level B RCE reports, as well as reclassification of CRs (e.g., no RCE required, downgrade from a Level B to a Level C, etc., as described in NP 5.3.9). 1. Presentation Bring copy of completed evaluation and any supporting information you may need to answer questions or to support statements or positions made in the RCE. If the evaluation was done by a team, members of the team should be present. Presentation slides are not required. Be brief and to the point when discussing the evaluation. Depending on the agenda, normally about 10 minutes is allotted to presentation and discussion of the RCE. The evaluators group head should attend the CARB presentation to support the evaluators conclusions and to accept additional actions as required.

2. Actions Recommended by CARB - CARB may accept a RCE as written, accept with comments, or reject. Rejections and comments resulting in significant changes to the RCE should be addressed in a revision to the RCE and reapproved. The revised (rejected) RCE should then be presented to CARB, preferably with as many of the same CARB members present as possible (at least, any that had significant comments). 12.0 EFFECTIVENESS REVIEWS Effectiveness Reviews are performed after corrective actions have been implemented to ensure the RCE identified and corrected root causes. This is a proactive assessment of the corrective actions versus waiting for an event challenge to determine effectiveness. The depth and duration of an effectiveness review should be commensurate with the significance and complexity of the problem. See Exhibit Q for general guidance and examples of when an Effectiveness Review might be appropriate.

Page 17 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT A PERSONAL STATEMENT Page 1 of 1

OEG 001 Revision 3 April 29, 1999

/ Name (Print) General Instructions: Date

/ Position Department

In your own words, describe your knowledge of the event facts, and your involvement in the event before, during, and after the final outcome. Include any pertinent verbal communications and specify who you spoke with (by name and/or position). Indicate the format of the communications (pre-job brief, direct assignment, inter-department interfaces, etc.), and who you spoke with. List any pertinent procedural or equipment conditions relating to the event. Use additional sheets as necessary. ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________ ___________________________________________________________________________________

Signature _________________________________ Page 18 of 52 INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT B EVENT AND CAUSAL FACTOR CHARTING Page 1 of 4 An event and causal factor chart (E&CF) is a graphic display of an event. The heart of the E&CF chart is the sequence of events plotted on a time line. Beginning and ending points are selected to capture all essential information pertinent to the situation. Often, failure modes that are not obvious become evident through this technique. E&CF charts are particularly useful for complex and complicated situations, and can be more useful than long narrative descriptions. They allow you to separate the many causal factors associated with complex events The E&CF chart graphically displays the relationship between the sequence of events, inappropriate actions, barriers, changes, causes, and effects. FORMATTING THE E&CF CHART All events (actions or happenings) that occurred during some activity - rectangles All conditions (circumstances pertinent to the situation) that may have influenced the course of events - ovals All events and conditions that are assumed or have not been confirmed - dotted line rectangles and ovals Primary effect(s) of a series of events (or inappropriate actions that may have led or contributed to the situation) - diamonds Causal factors (shape the outcome of the situation) - ovals shaded at one end (light) Root Cause - ovals shaded at one end (dark) Terminal event (end point of the evaluation, typically this will be the consequence of the event) - circle Other symbols may be used, as desired, to indicate barriers, broken barriers, process changes, or other items that contribute to the clarity. Provide an identification key for these symbols if used.

Page 19 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT B EVENT AND CAUSAL FACTOR CHARTING Page 2 of 4 CRITERIA FOR EVENT DESCRIPTION Events must precisely describe a SINGLE action or happening (quantified) and be based on VALID information (facts). Use a short sentence, usually with just one subject noun and one action verb. Chart scope should range from beginning to end of the situation sequence. Each event should be derived directly from the event and conditions preceding it. When this is not the case, it is an indication that one or more events or conditions are missing. Each event should be in the appropriate time relationship to the preceding and succeeding event.

Detail of the event sequence MUST be sufficient to ensure completeness of final report. HOW TO DEVELOP E&CF CHART STEP 1: Evaluate initial information and documentation What were inappropriate actions and/or equipment failures? When did they occur (during what task/evolution)? How did they occur? What were the consequences? Begin constructing the preliminary primary event line. Start early - use currently known facts Use yellow sticky notes. The events, factors, and conditions will probably need to be revised and rearranged. Define scope of chart from initial information. Initiating event, i.e., beginning point Terminal event, i.e., the reason for the investigation Add new information to preliminary chart. Events Primary directly leads to or follows a primary effect or inappropriate action Secondary impacts primary event, but is not necessarily directly involved in situation. Plotted on horizontal lines parallel to primary events line Conditions Initial During course of inappropriate actions or equipment failures After inappropriate actions or equipment failures

STEP 2:

STEP 3:

STEP 4:

Page 20 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT B EVENT AND CAUSAL FACTOR CHARTING Page 3 of 4 STEP 5: Identify failed barriers, changes, and causal factors. Analysis (Task, Change, Barrier, Cause & Effect, Interviewing) Ensure facts are validated and conclusions are supported by facts

REMEMBER There is no correct chart. The important thing to remember is to use the chart and the process to help discover the root causes and to convey that discovery process to others reviewing your investigation. The rules are not mandatory. Violate these E&CFC rules when it contributes to communicating the information. The intent of this process is understanding the sequence of events and the relationships of the conditions and causal factors.

Page 21 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT B EVENT AND CAUSAL FACTOR CHARTING Page 4 of 4

Electronic stencil for your E&CFC can be found in: j:\DATA\NP\oe\RCE\VISIO\Rce_e&cf.vss or \Rcechart.vss
Service Water Hydraulic Analyses Configuration RCE 97-032 EVENT AND CAUSAL FACTORS CHART Attachment 1

FSAR Section 9.3 Addresses CCW Design and Operation

Service Water System Flow Margin Limited

OI-71 First Issued 1987

WATER Program and Input Model Obtained from Bechtel


1989

SW Model Validated by Comparison of Analysis in Calc. 92-087 to Parameter Measurements


1992

Condition Reports 97-1777, 97-1690, and 94-633


1994-1997

Condition Report 971859


6/11/97

Calculations Assumptions Revised and Calculations Corrected

Documentation of Model Basis Limited

No Formal Training Provided on Model Development and Use

Model Basis and Number of Computer Runs Increasingly Complex

Calculations Completed Under Time Pressure

Inadequate Interface Between Engineering and Operations

Page 22 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT C TASK ANALYSIS Page 1 of 2

OEG 001 Revision 3 April 29, 1999

Task analysis is a tool that is used on evaluations where problems during performance of tasks contributed to the event. Performing a Task Analysis will provide the RCI with: 1. A clear understanding of how the task is normally performed. 2. Questions arising out of the analysis to be answered during the course of the evaluation, usually through interviewing. One of the first priorities when entering an evaluation is to understand as much as possible about the activity that was being performed. It may be necessary to obtain the required expertise on the team to be able to perform the task analysis. The task analysis will require a review of work documents, logs, technical manuals, and other documents in an effort to determine what the task is about and how it was to be performed. This process is called the Task Analysis method. 1. Paper and Pencil - the task is broken down on paper into subtasks identifying: Sequence of actions Instructions Conditions Tools Other materials associated with the performance of the task

This type of analysis consists of a review of logs, work documents, technical manuals, etc., to determine what the task was about and how it was to be performed. The steps, questions and concerns should be displayed on the preliminary event and causal factor chart. 2. Walk-Through - A step-by-step enactment of the task for an observer without carrying out the actual function. The observer makes notes of any differences between the actual performance enactment and the procedure steps. Personnel performing the walk-through should be people who actually do the tasks, but not people who were directly involved in the event. The walk-through should identify: How the task is "really" performed Problem areas such as: Discrepancies in procedure steps Human factors design in the man-machine interface Training, knowledge, or skill weaknesses Page 23 of 52 INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT C TASK ANALYSIS Page 2 of 2 Steps in Walk-Through Task Analysis:

OEG 001 Revision 3 April 29, 1999

1. Obtain preliminary information to understand what happened during the event. 2. Determine the scope of what is to be included in the walk-through. 3. Obtain necessary information: procedures, work package, etc. drawings interviews

4. Develop a guide for the walk-through to outline how the analysis will be conducted: identify key activities to be performed and observed identify activities to be recorded

5. Determine exactly what information is going to be recorded and how - one technique is to check off each step as it occurs. Discrepancies and problems may be noted in the margin or in comment space provided adjacent to the step. 6. Select personnel to perform the task who normally perform it. If a crew is involved, crew members should perform their normal role. 7. Perform the walk-through while observing and recording. Note any discrepancies or problems. Try to re-create the situation to obtain a sense of how the actual event occurred. The walk-through may be done in slow motion, stopping to address questions. The personnel performing the task may describe the activities from their perspective as they perform. The walk-through may be performed in real time to identify time-related problems. An actual task in the plant may be observed, but preparation as described above is necessary. A simulator or mock-up may be used.

8. Summarize and consolidate problems noted and questions to be answered during interviews. Identify possible contributors or causal factors for the event or failure. (1) Example of a Task Analysis Worksheet Steps in Procedure or (2) Walk through by Analyst (3) Questions/Conclusions Practice. (Enter step or trained individual. about how task number and short (State how actual matches was/should be performed. description.) procedure.)

Page 24 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT D INTERVIEWING Page 1 of 5 Interview Preparation

OEG 001 Revision 3 April 29, 1999

All interviews require preparation, no matter how simple the problem seems. Interviewing is a fact finding skill rather than a fault finding session. A. Develop a set of questions. The questions can be derived from the Events and Causal Factors Chart, Change Analysis, Barrier Analysis, and the enclosed Question Guide. Consider the preferred sequence of interviews. Make appointments. Select an appropriate location Allow time between interviews to reconstruct notes.

B. C. D. E.

Introduction/Opening The purpose of the introduction is to orient the interviewee and put him/her at ease. A. Explain the purpose of the evaluation and the interview (to identify what happened, how it happened, why it happened, and what can be done to prevent recurrence). Provide the interviewee with an overview of the material to be covered. Show interest and get the interviewee involved. Anticipate and answer the interviewees questions: What will happen with information (it will be used to determine root causes). Will my name be used (the report may include a list of interviewees). Why do you want to talk to me (we believe that you can help explain what happened)

B. C. D.

Page 25 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT D

OEG 001 Revision 3 April 29, 1999

INTERVIEWING
Page 2 of 5 Question/Answer The purpose of the interview is to obtain the interviewees recollection and understanding of the event. The following are some of the features of a successful interview. A. B. C. D. E. Begin with open-ended questions to allow the interviewee to provide his/her perspective. Listen carefully while taking notes. Do not interrupt. Keep questions short and to the point. Do not ask leading questions. Use primary questions (from the prepared list) to introduce a topic and use secondary questions to clarify information.

The Closing The closing accomplishes more than just concluding the interview. It provides an opportunity to validate information and obtain additional information. Summarize the information that was recorded. Set up the potential for a follow-up interview. Thank the interviewee for his/her help.

Page 26 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT D INTERVIEWING Page 3 of 5 Interviewing Guide A. Verbal Communications B.

OEG 001 Revision 3 April 29, 1999

Were instructions adequate and clear Were communication practices consistent Were plant communication systems adequate Were there problems communicating between work groups

Written Document Were documents complete, clear, and understood Were documents used for the task Were other documents referenced Were documents legible and current Were drawings, sketches, tables, etc., useable Were documents technically correct Were documents readily available Did the document contain appropriate prerequisites, initial conditions, precautions, cautions, and warnings Were problems with documents reported, resolved Was there any problem using the document to identify the correct unit, train, component Could the task be performed as required by the document

C.

Human Factors Were there any problems distinguishing/identifying components Were components labeled Were label identifiers consistent with work documents Were labels color-coded or otherwise readily apparent

D.

Physical Environment Was lighting adequate Were there housekeeping problems (water, oil, debris, etc.) Was there need to enter a confined space Was protective clothing available and used Was temperature/humidity a problem Was noise a problem Were there obstacles or distractions present

Page 27 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT D INTERVIEWING Page 4 of 5 E. Work Schedule F. How may hours had been worked prior to the event How much overtime had been worked prior to the event How many consecutive days had the person worked What time of day did the event occur When was the next day off scheduled to occur

OEG 001 Revision 3 April 29, 1999

Work Practices Determine procedure use (verbatim, guideline, not used, etc.) Were tools in good working condition Were all needed tools available Was self-checking used Was independent verification used Were radiological conditions understood Were system conditions understood Were short cuts used Were all the required people present

G.

Work Organization and Supervision Were duties distributed appropriately Was there enough time to prepare for the job Was there more than one simultaneous task Had the job been performed previously Were duties and responsibilities clear Was the supervisor at the job location periodically Were tasks coordinated among work groups Were priorities clearly established How long had this work crew worked together Was there an adequate pre-job briefing Were contingencies established for anticipated problems

Page 28 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT D INTERVIEWING Page 5 of 5 H. Training and Qualifications I.

OEG 001 Revision 3 April 29, 1999

Had the workers been trained to perform the task Was any training based on the actual task Did the worker have an understanding of the equipment involved Did the worker read and understand the work instructions Was any applicable training useful; qualified instructors Did training include mock-ups, simulator, etc. What were the differences between training and actual job How long since training was received Was sufficient time allowed for training How long since the task was last performed

Change Implementation Was there anything different since the job had previously been performed Were changes adequately reflected in procedures, drawings, training, labels, etc.

J.

Management and Administration Were there any policies, goals, or objectives that influenced the event Did the worker understand who he/she reports to Were roles and responsibilities clear Were quality requirements clear Is the expectation for problem identification and resolution clearly understood Was support adequate (procedures, training, engineering, planning, scheduling, radiological protection, clearance tagging, protective equipment, etc.) Were parts, materials, and supplies provided to support the job Was the reason for the job clear Was the job within the workers capabilities Were there unnecessary requirements Were there any conditions causing stress

Page 29 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT E CHANGE ANALYSIS Page 1 of 2 ALWAYS ASK -

OEG 001 Revision 3 April 29, 1999

What was different about this situation from all the other times the same task or activity was carried out without an inappropriate action or equipment failure? STEPS IN CHANGE ANALYSIS 1. 2. 3. Analyze the situation containing the inappropriate action or equipment failure. Analyze a comparable situation that did not have an inappropriate action or equipment failure. Compare the situation containing the inappropriate action or equipment failure with the reference situation. Write down all known differences whether they appear relevant or not. As the evaluation progresses, be alert to other differences that were not apparent during the initial review and add them to the list. Evaluate the differences for effect on producing the event. This must be done with careful attention to detail, e.g., a change in color or finish may change the heat transfer parameters and consequently affect system temperature. Integrate information relevant to the causes of, and contributors to, the inappropriate action or equipment failure into the investigative process via the E&CF chart.

4.

5.

6.

WHEN SHOULD CHANGE ANALYSIS BE USED? When causes of inappropriate action or equipment failure are obscure When you dont know where to start the evaluation When you suspect that a change may have contributed to the inappropriate action or equipment failure

Page 30 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT E CHANGE ANALYSIS Page 2 of 2 EXAMPLES OF CHANGES TO CONSIDER What -

OEG 001 Revision 3 April 29, 1999

operating parameters (i.e., changes in temperature, pressure, flow, cycle time, etc.) plant status, time of day, day of week, season of year, times when specific conditions exist (i.e., why does it work some times but not others?) physical location (i.e., why does it work in one location but not another?) how equipment is supposed to work (i.e., why does it work in one application but not another?) personnel involved (i.e., is one individual or crew using a different method or technique?)

When -

Where How -

Who

Example of a Change Analysis Worksheet Problem Statement: (Optional section. Write a brief statement of the event being analyzed and the question that needs to be answered.)

Previous Condition (or Error Free Condition, etc.)

Current Condition

Change / Difference

Impact or Assessment

(List all possible contributors one at a time, need not be in sequential order.)

(List comparable contributors.)

(List all differences without evaluation or value judgment or significance, whether relevant or not.)

(What effect did the change have on the situation.)

Page 31 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT F BARRIER ANALYSIS Page 1 of 3 DISCUSSION

OEG 001 Revision 3 April 29, 1999

Barriers are devices employed to protect and enhance the safety and performance of the plant. They can be physical or administrative in form. Barriers are erected to ensure consistent and desired performance of the plant. A single barrier is rarely relied upon. Generally, barriers are diverse and numerous - a defense-in-depth concept. Some examples of barriers commonly found in nuclear power plants highlight the importance of these devices as follows: PHYSICAL BARRIERS Engineered Safety Features Safety and Relief Devices Conservative Design Allowances Redundant Equipment Locked Doors and Valves Ground Fault Protection Devices Radiation Shielding Alarms and Annunciators Fire Barriers and Seals

ADMINISTRATIVE BARRIERS Plant Operating and Maintenance Procedures Policies and Practices Training and Education Maintenance Work Orders Radiation Work Permits Licensing of Operators Qualification of Welders Methods of Communication Certification of Health Physicists and Technicians Certification of Engineers Technical Specifications Regulations Supervisory Practices Work History

PLUS MANY MORE

Page 32 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT F BARRIER ANALYSIS Page 2 of 3 BARRIER TYPES

OEG 001 Revision 3 April 29, 1999

Barriers that promote (good design, labeling, work planning, procedures) Barriers that prevent (interlocks, locked doors, physical segregation) Barriers that discourage (caution signs, rope barriers, notes/cautions in procedures/briefings) Barriers that detect (holdpoints, checkoff lists, operator rounds, pre-job procedure reviews) Barriers that compensate (tests done at shutdown/low power, notifying control room prior to task)

BARRIER ANALYSIS METHODOLOGY 1. Identify target--Result of the event (e.g., Rx scram, ESF actuation, personnel injury, valve mispositioned, etc.). Target could also be desired result (e.g., successful completion of test). Identify a single hazard to the target--Typically start with the symptoms or failure mode(s) at the time the event occurred. This could also be energy applied to the system (e.g., monthly pump test). Identify all barriers. Integrate this information into the preliminary E&CF. Identify all apparent barriers that failed and allowed the event to progress. Determine HOW the barrier failed, e.g., the relief valve barrier failed because although the valve was functional the set point had drifted high. Determine WHY the barrier failed, e.g., the relief valve set point had not been checked since original installation because it is non-safety related. Validate the results of the analysis with information learned.

2.

3. 4. 5. 6.

7.

8.

Page 33 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT F BARRIER ANALYSIS Page 3 of 3

OEG 001 Revision 3 April 29, 1999

While barrier analysis identifies missing or defective barriers, it has a weakness. If the investigator does not recognize ALL failed barriers, the evaluation may be incomplete. Because using barrier analysis alone is very time consuming it is recommended that barrier analysis be used in conjunction with other techniques. When a RCA evaluation is initiated, you must think in terms of barriers. Naturally, the barriers established in plants differ widely and evaluation of them is dependent upon your knowledge. Regardless of variations in barriers at plants, RCA provides the framework for barrier assessment because it focuses on precise barrier categories that have proven to be critical in identifying equipment failures. Corrective actions from RCA evaluations usually include modification of existing barriers, but caution should be taken before considering additional barriers so that additional failure modes are not introduced. Example of a Energy(Hazard)/Barrier/Target Analysis Worksheet Energy/Hazard Barrier Assessment Target Monthly pump test Procedure No step to open Successful completion of discharge valve. test Operator New Operator. Did not QV&V or STAR Supervisor No oversight of first time evolution.

(List one at a time, need not be in sequential order.)

(Identify all applicable (Identify if barrier was physical and missing, weak, or administrative barriers ineffective and why.) for each consequence.)

(Identify all applicable targets such as individual organizations, equipment, facilities, and processes.)

Page 34 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT G FAILURE MODES AND EFFECTS ANALYSIS Page 1 of 3 THE FM & E ANALYSIS PROCESS 1. Develop a list of possible failure modes. Possible sources or references to develop the list might include the following: Previous failures from equipment history data bases Known failures from industry user groups Known failures from original equipment manufacturer Previous failures from other stations Failure diagnostic programs, guides, and tools (EPRI ERCAWS, computer aids, consultants, etc.)

Possible failure modes can be documented on a Fishbone Diagram where each failure mode is a major rib of the fish. As an alternative, the major ribs can be general categories such as Human Performance, Procedures, Equipment, and Facilities. 2. Collect physical evidence (NOTE: this can be performed concurrently with Step 1.) Physical evidence should be gathered to completely understand the WHAT and HOW of the failure.

CAUTION Care must be taken while gathering evidence not to accidentally destroy other evidence. For example, if a component must be disassembled, care must be taken to capture all "as found" conditions. Do not clean or contaminate fracture surfaces. Measurements, photographs, video tape, or other methods should be considered to preserve evidence.

Evidence should be gathered to validate or refute the postulated failure modes. For example, if one of the postulated failure modes is WATER HAMMER, then conduct a system walkdown to look for evidence of water hammer, such as damage to small pipe or instrument connections, etc.

3. Evaluate each possible failure mode against the physical evidence to validate or refute it. In other words, determine whether the failure mode would have produced the physical evidence that exists.

Page 35 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT G FAILURE MODES AND EFFECTS ANALYSIS Page 2 of 3 4. Continue Steps 1 through 3, and through a process of elimination reduce the list to the single failure mode or the most probable failure mode(s). 5. Evaluate the single or most probable failure modes using the "Cause and Effect" process to determine the root cause(s). HINTS ON USING THE FM & EA PROCESS A thorough understanding of the failed equipment is necessary in order to conduct FM & EA. A highly knowledgeable subject matter expert is needed. If the evaluation team does not possess a high level of knowledge, an expert needs to be recruited from elsewhere inside or outside of the organization. Possible failure modes should not be ruled out until physical evidence validates that it should be eliminated. The evaluation may need to look for a lack of evidence to eliminate a particular failure mode. The process may need to be repeated to identify intermediate failure modes until the primary failure mode is determined. Examination of physical evidence may need to be performed under laboratory conditions. If that is the case, it is important to get laboratory personnel involved as early as possible. It is highly recommended that laboratory personnel visit the location of the failure to understand layout, environmental conditions, history, etc., that may have contributed to the failure. If the component failure was catastrophic, physical evidence may have been lost or destroyed in the failure (for example, electrical insulation is destroyed by fire). If that is the case, other similar components can be examined. Also, possible corrective actions to consider are methods to capture and preserve physical evidence in future failures.

Page 36 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT G FAILURE MODES AND EFFECTS ANALYSIS Page 3 of 3

Environment

Natural Maintenance Bird Feces Salt Spray Artificial Washing Line Washing After Outages CB-EG Transmission Insulation Failure Degradation Contamination LineSectionalizing Line Failures Sugar Cane Burning Smog Contamination

Design

Operation

Page 37 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT H CAUSE AND EFFECT ANALYSIS Page 1 of 1 CAUSE AND EFFECT PRINCIPLES All events are caused to happen. These events are the result of plant conditions, plant design, human performance, etc. A bond/relationship exists between cause and effect. Root cause(s) can be found by examining the relationships. Ask why?, usually five to seven times to determine root cause.

HINTS ON USING CAUSE AND EFFECT ANALYSIS Often causes and effect analysis will lead to management-controlled root causes (also called Organizational and Programmatic causes). When more than one cause is responsible for an effect, each cause must be evaluated. Cause and effect is most effective when used within the framework of the E&CF chart. It is not a stand alone method because the situation must first be unraveled to the point where ALL failure modes are identified. This is particularly true in situations involving multiple failures. This process of cause and effect provides a logical, structured guide to maintaining the evaluation on track, but will require good judgment and experience to be effective.

Repeat Cause and Effect Analysis Until: 1. 2. 3. 4. 5. The cause is outside of the control of the plant staff The cause is determined to be cost prohibitive The equipment failure is fully explained There are no other causes that can be found that explain the effect being evaluated Further cause and effect analysis will not provide additional benefit in correcting initial problem

Page 38 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT I PARETO ANALYSIS Page 1 of 1

OEG 001 Revision 3 April 29, 1999

Purpose & Description- Pareto Analysis is the process of determining the Vital Few factors responsible for a particular problem. A Pareto Chart is a bar chart of failures ordered by frequency of failure, cost of failure or contribution to system unavailability.
E D G " A " H o u rs U n a v a ila b le
5 00
95 %

E D G " B " H o u rs U n a v a ila b le


100 % 5 00
94% 98%

100% 90%

4 50 4 00
7 4% 82%

89%

90% 80% 70% Hours Unavailable

4 50 4 00 3 50
66% 81%

80% 70% Percent 60%

3 50
Hours Unavailable

3 00 2 50 2 00 1 50 1 00 50 0
13 3 0% 9 127 246 48%

63%

60% 50% 40% 30%


Percent

3 00 2 50 2 04 2 00 1 50 1 00 50 0 1 30%34 1 07 1 02 89 26 16
50%

50% 40% 30% 20% 10% 0%

89 64 53 48 44

20% 10% 0%

1. Determine the Effect or Problem. What is the problem to be addressed? For example, is system reliability or availability of concern (or both)? 2. Decide how the Effect should be measured. Determine how to measure the problem. For example, frequency of failure is used to measure reliability; duration of failure is used to measure availability. 3. Decide how the Effect can be stratified. Failures are typically stratified by system equipment or component, although other strata may be used. 4. Interpret the Results. What does the Data Reveal? Which failures or causes are the Vital Few? Pareto Chart Construction 1. 2. 3. 4. Collect the data and group the events by category or strata. Order the categories from highest to lowest (frequency, unavailability, etc.). Draw bars for each category; the bar heights equal the categorys frequency/duration. Develop the cumulative line, adding the impact of each category from left to right.

Page 39 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT J TROUBLESHOOTING / FAILURE ANALYSIS Page 1 of 1 Purpose & Description - Troubleshooting is the diagnostic process of determining which component(s) failure resulted in the observed system failure. Failure Analysis determines why the component failed (physical failure mechanism). Troubleshooting and Failure Analysis comprise the analysis portion of a root cause analysis. Process - Troubleshooting/Failure Analysis steps include: 1. Determine the Failure Sequence/Circumstances. Gather initial data relevant to the failure event. This includes failure reports, operations logs, strip chart recordings, etc. 2. Develop a Troubleshooting Plan. Develop a plan to diagnose the failure. Determine which subsystems/equipment/actions could have caused the failure, develop diagnostic tests that will eliminate or confirm the potential failure sources. 3. Identify the Failed Part. Perform the diagnostic activities identified by the plan. Identify the failed part or parts. 4. Confirm Failure of the Part. (When laboratory analysis is required) Confirm that the part declared to have failed by the troubleshooting process is, in fact, failed. 5. Develop a Failure Analysis Plan. Develop a plan to determine why the part failed. This may included laboratory analysis, process analysis (procurement, maintenance, etc.). 6. Analyze the Parts Failure Causes. Perform the failure analysis tasks outlined in the plan. Determine the physical causes of failure. 7. Determine the Sources of these Causes. Determine if the part is capable of performing its mission. Determine which management system failed and requires modification in order to prevent the reoccurrence of the failure. 8. Develop a Conclusion & Recommendations. Review the facts and data, draw conclusions regarding the part(s) which failed, the physical cause of the failure and the processes which resulted in the failure. Develop corrective action recommendations.

Page 40 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION EXHIBIT K FAULT TREE ANALYSIS Page 1 of 1

OEG 001 Revision 3 April 29, 1999

Purpose & Description - Fault Tree Analysis develops a logic model that is used to identify and verify possible causes of failure.
EXAMPLE FAULT TREE Mechanical Pump Seal
Cooling Water Pump Seal Failure

Primary Seal Failure

Secondary Seal Failure

Carbon Face Failure

O-Ring Failure

Carbon Face Wear

Loss of Cooling to Seal

Excess Pump Vibration

Loss of Cooling to Seal

O-Ring Wear

Improper Installation

Thrust Bearing Misaligned

1. Determine the Failure (Top Event) to be analyzed. 2. List High-Level events (subsystem or functional failures) which could have caused the failure event. 3. Proceed to determine how the high-level events could have occurred. 4. Identify basic events (equipment, component, or part failures, human errors) which could have caused the failure. 5. Develop and implement diagnostic tests to exclude potential causes or verify causes.

Page 41 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT L DEVELOPMENT OF RECOMMENDATIONS FOR CORRECTIVE ACTIONS Page 1 of 1 Consider corrective actions to address the following: root causes (prevent recurrence) failure modes (repair what is broken) symptoms (detect future degradation before failure) common mode failures (other components, train, systems, unit, sites, departments, programs, etc.) effectiveness follow up (are actions effective) Recommendations need to: address issues be cost effective be within control of site meet or exceed industry standards Discuss proposed corrective actions with people that have to implement them. Get their input, suggestions, and buy-in. Factors to consider: cost risks and/or consequences of actions or inaction (new failure modes) mitigation or prevention if addressing root cause is cost prohibitive Constraints to consider: time (short-term vs. long-term, temporary fix vs. action to prevent recurrence) resources political realities Understand the following: requirements (CFR, codes, FSAR, etc.) commitments goals and objectives previous or similar situations vested interests IMPORTANT! Avoid the shotgun approach. Excessive and unnecessary corrective actions not only add burden to staff, but introduce the possibility for new failure modes. For proposed corrective actions, ask which root cause they will address. If they do not address a root cause, are they needed?

Page 42 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT M EVENT EVALUATION AND ROOT CAUSE ANALYSIS TECHNIQUES APPLICATION GUIDELINE Page 1 of 1

Equipment Failures Failure Modes & Effects Analysis Fishbone Diagram Pareto Analysis Problem Solving Quality Improvement Processes Statistical Process Control General Error Model System Task Analysis Events & Causal Factors Charting Change Analysis Barrier Analysis Interviewing Cause & Effects Analysis Stream Analysis X X X

Human Performance Problems

Organization & Programmatic Problems

Trend X X X X

X X X X X X X X X X X X X X X X X X X X X X X

X X X X X

Page 43 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT N RCE QUALITY INDEX SCORE SHEET EXAMPLE Page 1 of 2 RCE QUALITY INDEX SCORE SHEET EXAMPLE QUESTIONS AVG OEG 001 DESCRIPTION SCORE REFER. Relevant Facts & Data Quantitative information used to develop 6.4.1 Covered? Y 1.00 E&CF chart, timeline, process map, etc. All statements are qualified and quantified (Q&Q), not based on opinion or assumptions. Critical Data The critical data used is verified with an 6.4.1 Independently Y 0.65 independent source of data (QV&V techniques Verified? used). Independent source of data could be through use of interviews, surveys, etc. Root Cause to Isolated The appropriate issues have been isolated as 6.5.4, HE/O&P/MI? the root cause. It is possible that the root cause 6.5.5, Y 0.65 could be found to be an isolated human error 6.6 (low incident rate or high individual rate) or programmatic or management issues, but this conclusion could be easily mistaken. All Possible Failure All possible failure modes need to be identified 6.5.1, Modes of HE/O&P/MI Y 0.75 to adequately perform a Stream Analysis to 6.5.2 Covered? identify the root cause. Reasons for excluding some may be appropriate. Corrective Actions Corrective actions too narrowly focused, 6.6, Too Narrowly N 0.90 unless appropriate, may not correct the root 6.7, Focused? * Exhibit L cause such that similar events are prevented. Corrective Actions to Prevent Recurrence (CAPR) should address each root cause. Corrective actions should meet four criteria (*). Is Failed Barrier or Mixing failure modes, failed barriers, and/or 6.5.4, Failure Mode Treated N 0.60 inappropriate actions with, or identifying them 6.6 as Root Cause? as, root causes may result in misdiagnosis. O&P/MI Root Causes Benchmark root causes against internal and 6.4.2, proven substandard industry standards. Standards might be found 6.6 through quantitative Y 0.60 from INPO good practices, OE information, benchmark analysis? Nuclear NETWORK questions, etc. Cannot meet or exceed industry standards if this is not identified. Root Cause relies on O&P issues should be verifiable through 6.5.3 interview statements N 0.90 factual information; interview statements are to identify O&P used to confirm. issues?

Page 44 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT N RCE QUALITY INDEX SCORE SHEET EXAMPLE Page 2 of 2 RCE QUALITY INDEX SCORE SHEET EXAMPLE QUESTIONS AVG OEG 001 DESCRIPTION SCORE REFER. Previous similar Using a single event to reveal and diagnose 6.4.3, events considered? Y 0.75 global O&P issues may result in mis-diagnosis. 6.4.4 & An explanation of search criteria and results Note can support conclusions related to extent of condition. Why were previous Previous corrective actions may have been too 6.4.5 corrective actions for Y 0.35 limited, inappropriate, counter-productive similar events (i.e., caused more errors), etc., or they may not ineffective? have been implemented. Exhibit B Other items from OEG 001: TOTAL SCORE 10 7.15 6.3.3 Events & Causal Factors Chart 6.4.5 & Extent of Condition Note, 6.4.7 *Four Criteria: 1. 2. 3. 4. Will immediately correct discrepant condition and prevent concurrence. Can be implemented by reasonable action. Will meet or exceed industry standards. Is cost effective. Reviewer: /

Page 45 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT O ROOT CAUSE EVALUATION PLANNING GUIDE Page 1 of 1


Working Days Day 1 Day 1 Varies Calendar Day ______ ______ ______ 1. 2. 3. Process Review CR for issues to be addressed in RCE. Talk with Issue Manager to determine scope of evaluation and personnel involved. Determine information gathering techniques: interviewing, change analysis, barrier analysis, photography, etc. Gather and QV&V information. Develop E&CF chart. Timing varies, and chart may be continuously revised, but complete rough draft E&CF prior to drafting RCE report. (May be developed before or during interviewing.) (RCE Coordinator can computerize for you, if needed.) Develop interview questions. Review as needed with RCE Coordinator, CAP Manager, or Issue Manager: plan, questions, interviewee list Set up and perform interviews. Complete other information gathering. Requests for information via Nuclear NETWORK, etc., should be made as early as possible. Develop draft report. Perform searches on NUTRK, CHAMPS, etc., to identify previous events, OE and industry standards, etc. Analyze information and determine pertinent facts. Compare facts against acceptable standards to identify unacceptable conditions or performance. Identify inappropriate actions and equipment failures. Use barrier analysis, etc., to determine failure modes, conclusions, and causal factors. Identify root cause(s) and corrective actions to prevent recurrence. Identify extent of condition. Is condition fixed? Determine if there are generic issues associated with this event and whether they should be addressed in this RCE or by a separate CR.

Day 1 Day 2 Days 2-6

______ ______ ______

4. 5. 6.

Day 7-11 Day 7-11

______ ______

7. 8.

Day 11 Day 12 Day 13

______ ______ ______

9.

10. Validate root cause, pertinent facts, and corrective actions through interviews with persons not involved in event, scenarios (is similar event prevented), etc. 11. Provide draft report to interviewees and personnel who will be performing corrective actions. Provide copy to the RCE Coordinator for quality review and comment. 12. Revise draft report as necessary, based on input from the reviewers. RCE Coordinator, CAP Manager, or CARB can provide assistance in resolving disagreements, if necessary. 13. Provide revised draft to appropriate managers for approval. 14. Provide signed report to RCE Coordinator for distribution. 15. Close the action item in NUTRK. Initiate new action items for corrective actions from approved report.

Day 18

______

Day 19 Day 21 Day 22

______ ______ ______

Page 46 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT P ROOT CAUSE ANALYSIS PACKAGE COMPLETION CHECKLIST Page 1 of 2 Data Sources _____ Documentation required to support factual statements is included or referenced to retrievable documents.

Evaluation Sources _____ The RCE manual was used to determine which method of causal factor analysis is most appropriate for each category of event. (More than one method may be necessary to fully analyze the event or condition.) Documentation (e.g., worksheets, checklists, statements, copies of stripcharts) is referenced in the report and attached to confirm the method used to determine the root causes and causal factors and make a complete record.

_____

Method(s) - Check method(s) used to determine root cause(s) _____ _____ _____ _____ _____ Scope _____ Scope of evaluation addresses CR problem statement(s) and/or the issue managers direction. Event & Causal Factor Chart _____ Human Performance Analysis Fault Tree Analysis _____ O &P Failure Analysis Change Analysis _____ Equipment Failure Analysis Barrier Analysis _____ Task Analysis Other (describe): ____________________________________________

Benchmarking/Extent of Condition _____ _____ Internal and external operating experience have been checked for adverse trends/conditions. INPO Nuclear NETWORK or other sources have been used to identify industry standards.

Validation _____ Root causes and contributing factors were tested for validity.

Page 47 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT P ROOT CAUSE ANALYSIS PACKAGE COMPLETION CHECKLIST Page 2 of 2 Corrective Actions _____ Each root cause has a recommendation aimed at eliminating or minimizing its recurrence. (Contributing factors have corrective actions if appropriate.) Corrective Actions to Prevent Recurrence (CAPR or CATPR) are clearly identified. The appropriate group for implementation is identified for each corrective action. All corrective actions have been entered into NUTRK for followup.

_____ _____

Report/Communications _____ _____ _____ _____ _____ _____ The report is in the format recommended in Section 7.1 of this manual. Peer review has been completed. MSS presentation has been completed (Level A CRs only). Report has been submitted to RCE coordinator for quality scoring. Report has been approved by the appropriate group head. Report has been submitted to RCE coordinator for final distribution.

Page 48 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT Q EFFECTIVENESS REVIEWS Page 1 of 2 Effectiveness Reviews are performed after corrective actions have been implemented to ensure the RCE identified and corrected the root cause(s). This is a proactive assessment of the corrective actions versus waiting for an event challenge to determine effectiveness. The depth and duration of an effectiveness review should be commensurate with the significance and complexity of the problem. The following provides general guidance: 1) The following are considerations for electing to perform an effectiveness review: type of cause (isolated human error vs. programmatic issue; localized issue vs. widespread issue) type of corrective action (corrective action to prevent recurrence vs. broke-fix, improvements) potential impact on the business unit should another event occur 2) An effectiveness review should ensure the following: all corrective actions to prevent recurrence are complete the condition and cause(s) were corrected no additional corrective actions are required the corrective actions did not create any new adverse conditions 3) The effectiveness review can be performed using various techniques including: field verification or observation audit surveillance self-assessment or assessment from outside entities, mini- (focused) assessment survey (formal or informal) records (e.g., logs, CRs) review personnel interviews testing 4) Effectiveness reviews will be tracked in NUTRK under a CARB TRACK ID. 5) Completed Effectiveness Reviews should be presented to CARB as requested. 6) If corrective actions have not been effective at removing the root cause or condition, one of the following actions may be appropriate or may be directed by CARB: initiate a CR to document the ineffective corrective action revise the root cause initiate new corrective actions

Page 49 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT Q EFFECTIVENESS REVIEWS Page 2 of 2 Example: 1) A root cause analysis was performed to determine why appropriate LCOs were not entered during an event. The root cause analysis determined that there was a lack of knowledge and understanding by the operators of the basis and requirements of certain technical specifications (knowledge based error on the part of several operators). The corrective action to prevent recurrence was to provide training on the basis and requirements of the technical specifications in question, and provide practical application in the classroom and simulator. An effectiveness review could be performed to determine if the classroom and simulator training adequately resolved the knowledge and application deficit by administering a test in re-qualification training after the corrective actions have been completed. 2) A root cause analysis was performed to determine why non-QA parts were installed on a QA component. The root cause analysis determined that the planner writing the work plan did not understand the method used at PBNP to determine quality classification of sub-components or class breaks (green lines). The root cause also determined that the problem was widespread in the organization, that green lines were rarely used, and that the green lines were not readily accessible to the planners. Corrective actions to prevent recurrence included placing the green line prints in the area, training on their use, and establishing clear expectations for their use. In this case, an effectiveness review could be performed to determine if the work practice of the planners has been favorably impacted after completion of the corrective actions by performing a mini-self assessment by monitoring, observing, and discussing determination of quality classification with the planners. Note that in both examples: Corrective actions to prevent recurrence were targeted for the effectiveness review. The conditions were not due to isolated human error. The effectiveness review is proactive; i.e., it is focused at measuring the effect of the corrective actions before the organization is challenged.

Page 50 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT R APPARENT CAUSE DETERMINATION GUIDELINES Page 1 of 2 An apparent cause analysis is a brief evaluation that is expected to take an individual familiar with the issue a few hours to complete. In contrast a formal root cause analysis normally takes many person days to complete. Formal use of root cause methodologies is not required. Some investigation and analysis to support definition of the adverse condition, its apparent cause(s) and its resulting corrective action(s) is required. In other words, a description of what happened, what was looked at, and what was found should be included in the report. An apparent cause determination is a relatively brief evaluation which serves 4 main functions: 1. Ensures the adverse condition is understood and corrected. This is accomplished by fully describing the condition (when, where, what) and taking remedial (corrective) action to restore the adverse condition to an acceptable state. For Conditions Adverse to Quality, the corrective action is required by 10CFR50, Appendix B, Criteria 16. For other adverse conditions (e.g. industrial safety, unacceptable economic consequences, etc.) corrective action is required by management expectation. 2. Ensures that Significant Conditions Adverse to Quality are not present. CRs are assigned a significance level and evaluation method based on the event/condition description provided by the initiator and his/her supervisor. This decision is based on limited information. Documenting a review of the extent of condition, generic implications, and/or potential for common mode failures provides additional assurance that a Significant Condition Adverse to Quality was not inadvertently assigned as an apparent cause evaluation. If the evaluation indicates that there are serious issues which warrant an increased level of investigation, it is expected that the evaluator will discuss the issue with appropriate management. Issues uncovered which involve operability and/or reportability should be immediately reported to the Shift Manager. 3. Identifies apparent causes (failure modes) for trending purposes. Identification of root cause for CRs assigned an apparent cause evaluation is not required nor is it expected. Identification of the apparent cause or failure mode is expected. This information is used for trending purposes to identify common failures and common causes which allows analysis and correction prior to the occurrence of a significant event. If an apparent cause cannot be determined in a reasonable time, it is better to indicate that the apparent cause is unknown than to guess.

Page 51 of 52

INFORMATION USE

NUCLEAR POWER BUSINESS UNIT OE GROUP GUIDE ROOT CAUSE EVALUATION

OEG 001 Revision 3 April 29, 1999

EXHIBIT R APPARENT CAUSE DETERMINATION GUIDELINES Page 2 of 2 4. Allows consideration of corrective actions to prevent recurrence. Apparent cause determinations are much less extensive than root cause evaluations and normally stop at failure mode identification. They typically may not accurately identify underlying root causes. Corrective actions to prevent recurrence should be considered carefully to ensure that corrective actions to prevent recurrence based on apparent cause do not become ineffective barriers which may actually increase problems in the future. In considering the "risk" associated with specifying corrective actions to prevent recurrence, both the certainty of understanding the underlying cause and the potential detrimental effects of implementing the corrective action should be evaluated. The evaluation should include: Description of occurrence (dates, times, inappropriate actions, failures, immediate actions, etc.) if this information is not clear or thorough in the condition report. Analysis (what was reviewed, who was interviewed, other data) Extent of condition (previous events, generic implications, etc.) Conclusions (apparent cause, failure mode) Remedial (corrective) actions, responsible manager, and due date (if not already complete) Corrective actions to prevent recurrence, responsible manager, and due date (if applicable) Appendices, Attachments, supporting documentation (if applicable)

Page 52 of 52

INFORMATION USE

Você também pode gostar